Pixar Makes Painterly CG: New Research Could Change The Look of Their Films

Last summer at SIGGRAPH, Pixar presented a paper offering some clues about one of the major new directions that CG feature animation is headed. The paper, “Stylizing Animation By Example,” explored how filmmakers could achieve more expressive rendering styles that disregard the perfect boundaries of computer graphics rendering and mimic traditional painting techniques.

While non-photorealistic rendering is not a new concept, the system devised at Pixar contains a level of automation and consistency that would allow a sophisticated painterly style to be scaled for feature film production. According to Michael Kass, a senior scientist at Pixar, there are two major innovations in the project: 1) temporal coherence, which is the fluidness of images from frame to frame, and 2) the control over the final look that is gained by an artist who paints individual keyframes. The full paper, as well as additional documentation, can be downloaded on Pixar’s website.

Another Pixar senior scientist Kurt Fleischer, who was the technical director of the project, explained its genesis to Computer Graphics World: “Michael [Kass] was working on image filtering when some people in production became interested in the artistic images…We experimented with different kinds of techniques that looked nice and continuous over time, but we didn’t have artist-specific control. We only cracked that in the last couple of years.”

Fleischer doesn’t identify the Pixar artists who were interested in the approach, but it bears noting that the director and art director who created the artwork for the research project were Teddy Newton and Don Shank respectively. Veteran Pixar animator Sanjay Patel created the animation of the skater. Other people credited in the paper are character shading TD Jamie Frye, production designer Ralph Eggleston, and animator Angus MacLane.

With this new technique, an artist can paint 2D illustrations over a handful of the CG keyframes (“one every 10 to 20 frames”), and the software will fill in the hundreds of inbetweens while maintaining temporal coherence and the desired painterly style. The move away from photorealism has been a longstanding goal of animation artists. David Gainey directed Fishing at PDI/DreamWorks in 1999, and more recently, John Kahrs won an Oscar for Disney’s Paperman, which explored a similar set of ideas as the Pixar paper. All kinds of rumors are being tossed around currently about upcoming Disney features that will use non-photorealistic rendering styles, and my guess is that we’re 3-5 years away until a major studio feature is released using such techniques. As for Pixar, they’ll most likely create an animated short using these rendering techniques before incorporating the technology into a feature-length project.

The elephant in the room that needs to be addressed at some point is whether changing the style of CGI to mimic traditional media is a step forward or backward for the art of computer animation. As much as some computer artists desire to eliminate the digital origins of their work, a clear and troubling division exists between CG processes devised to replicate traditional media, and a film director like Aleksandr Petrov who uses actual oil paints to create animated films. When Petrov animates in oil, the plasticity of the materials directly informs the movement of his figures. In the Pixar process, the decorative surface style is incongruously layered atop a pre-existing piece of animation. In other words, the use of different media in computer rendering doesn’t result in different animation effects because the animation is already created; it’s an utterly arbitrary aesthetic embellishment.

This dissonance between visual technique and animation remains central to the aesthetic discord in non-photorealistic rendering. While there is a certain empty attractiveness to rendering in an illustrative style, the surface seductiveness has no influence on either the film or the animation besides reducing the emotional authencity of the artwork by distancing it further from its digital roots. Perhaps that’s the goal, but it seems to me an aesthetic dead-end.

Dead-end or not, research continues apace into these rendering techniques and its impact will soon be felt in feature animation. This most recent Pixar research into stylized rendering was conducted by the following individuals, who include both Pixar employees and graduate students/PhD candidates from Stanford University, University of Toronto and University of Washington: Pierre Benard, Forrester Cole, Michael Kass, Igor Mordatch, James Hegarty, Martin Sebastian Senn, Kurt Fleischer, Davide Pesare, and Katherine Breeden.

(Thanks, Jonah Sidhom, via Cartoon Brew’s Facebook group)


  • Delta

    Something just looks…kind of off. I mean, it looks cool, but there has to be a middle ground. In all, I think it depends on what kind of movie they’re trying to make- what kind of story and characters it has. The medium has to work for the story and its environment.

    • Steve

      Yeah, I brought the same issue of it looking off when some friends were showing this on facebook. I think it’s the lack of motion blur, and it makes me wonder what the possibilities are for drawing blurs and smears with this new tool.

  • anakinbrego

    Can we just bring back hand drawn animation please!

    • Radio

      because drawing everything by hand is so very quick and convienient -.-’

      • foljs

        Because “quick and convenient” is what matters when creating art… -.-’

        • Aaron R.R.R. Nance

          I believe I agree with you however I would qualify that slightly and say that “quick and convenient” really matters when there are so many other aspects involved in the medium; anything that produces quality results and saves time on one aspect frees the artists up to be more creative and less technical. That is a good thing. Photoshop and Wacom tablets are technological aids that lower the barrier for artists trying to convey their ideas. Sure they can be abused to create lazy inorganic work but used properly they are capable of gorgeous effect.

          I’m excited to see where Pixar takes this.

    • Strong Enough

      Can’t we have both please?

      CG is a gift to people who can’t draw but want to be animators

      • Alex

        Just because people work in 3D, doesn’t mean they can’t draw and ‘want’ to be animators. They are animators, and they often can draw very well too. CG refers to computer generated, & you’ll find that a lot of 2D productions also qualify as this, nothing is really hand drawn on paper anymore.

        • Strong Enough

          what? I never said that all of them can’t draw. But plenty of 3D animators don’t have the talent to draw but still want to be animators. this allows them to follow there passion.

          • Chris Sobieniak

            That wouldn’t surprise me if that’s all it did.

          • Strong Enough

            what wouldn’t surprise you?

      • AbdielUngo

        most of them, but it doesn´t apply to all

      • kevin

        I don’t believe people can’t draw but can 3D animate. I couldn’t draw, tried 3D animation for a year and failed miserably. I took my mentor’s advise and took many drawing classes (ex. figure drawing, anatomy, gesture drawing, etc.) went back to 3D animation and it looked much
        better.

        I heard from friends who actually worked in the industry that there were incidents like an animation sequence hand-drawn by a 2D animator was passed on to a 3D animator, and the 3D animator moved 2D into 3D. Basically, the 3D animator used 2D animation references from the 2D animator to animate 3D. I guess that 3D animator couldn’t draw, and what if there was no 2D references.

        Somebody need to prove me wrong with solid examples that 3D animators who can’t draw but can animate marvelously.

        • Mike L

          Plenty Dreamworks animators have already stated that they are not good draftsmen. If you need proof try watching a few DW features and see if they can animate or not.

      • Animator606432

        I agree, why can’t we have ALL styles of animation? Nobody, even old cranky traditionalist want just one style of animation round. Variety is the spice of life and that especially is true in art. I’m always nervous about these things, to be honest, because I’m worried by chosen craft will become obsolete if a computer can do it faster and better. It has happen before (cell painters) so I don’t see why it wouldn’t happen again.

        At the same time, It doesn’t mean I hate CG. I love it, I could never do what those animators do, so I’m glad so them experimenting with different styles.

  • Jason

    French animators are already doing much prettier, cooler work compared to this Photoshop filter effect lookalike.

    Salesman Pete, League of Legends Get Jinked videos for starters.
    http://vimeo.com/user2825651
    http://www.youtube.com/watch?v=0nlJuwO0GDs

    I mean it’s that classic effect of hand done artistic quality ALWAYS looks better than automated trash. You can’t shortcut this stuff which is weird as Pixar originally held strong to the same theories in their animation. I hope they realize it holds true for visuals as well.

    • alt animation podcast

      Most Supinfocom student shorts have this render technique and it makes them stand out in my eyes!

    • IJK

      Think you might have missed this part:

      “While non-photorealistic rendering is not a new concept, the system devised at Pixar contains a level of automation and consistency that would allow a sophisticated painterly style to be scaled for feature film production.”

      Many shorts have been made with gorgeous painted styles, but that’s the thing, they’ve all been shorts. Maybe they were shorts because doing that same technique for 180 minutes + however much scrapped footage the average film would be unbearably tedious. Pixar wants to lessen the time it takes for that technique but with the same results.

      • Jason

        Stop motion animation and anime feature disagree with your assessment.

        • IJK

          Nightmare Before Christmas took around 3 years. Then they improved techniques so things like Coraline could be made in 1.

          Princess Mononoke took two years. Technology gets better, faster to make complex visuals.

          Toy Story 1 took roughly four years to make. Guess what, technology got better, now they can crank out How To Train Your Dragon in a year and a half.

          Before that, 2D features took a hell of a long time to make. Technology gets better.

          How exactly are we suppose to get fast at making these complex features if we don’t perfect the technology first? People didn’t just start out making these films super fast. And those films I listed were in the 90′s, not that long ago.

          • barney miller

            Coraline took 3 years to make. I know because I worked on it.

            Many more years were spent on How to Train Your Dragon. Again, I know because I worked on it.

            Btw, why are you so interested in making these films quickly? What’s the rush? Things made from scratch take longer to make no matter what the technique is.

          • John

            There’s not really a “rush” to make them, but studios often don’t want to put their money into a technique that would take two to three times the amount of time and effort when people will go see their films even if they continued to use the same aesthetic over and over again.

            Example: Pixar is not going to invest the money in all watercolor-animation where each cel is painted by hand because it would take just way too long. HOWEVER, if they can achieve that look with their software in the same amount of time it takes them to make any of their other films, they’ll totally be up for it.

            Isn’t that why stop-motion has sort of died off (And Laika is attempting to revive it)? Because 3D is seen as a more appealing alternative? People complained that the look was nowhere near as good as stop-motion, so people try to get 3D to look as close to stop-motion as they can. That’s when we get things like “Room on the Broom”.

            Why don’t they just do it in stop-motion instead of trying to replicate a look? Because sometimes studios don’t want to invest in a medium they think will fail. And sometimes they only deal in a certain medium (I don’t think Pixar will be making a stop-motion film anytime soon) but they are still inspired/influenced by the look.

    • mmcfee

      “automated trash”

      Gee. Why not just call them “poopy heads”. If you’re going to throw around juvenile generalizations, get some poop in there. All the other kids are doing it.

      Seriously. If a Pixar animator were standing in front of you, would you call his work “trash”? Or only on the Internet?

      • Jason

        Notice how I didn’t mention the animation? It’s the render. Your anger is misplaced.

    • George_Cliff

      I really don’t see that video clip making your point. Yeah there is some nice VFX animation but the vast majority, the humans and everything else is drek.

  • alt animation podcast

    The animation plays to the camera in a very 2d way, which I think has more impact than the rendering itself

  • J

    Lets just be thankful that they’re exploring and experimenting with rendering instead of finding more ways to make hair believable.

  • http://dailygrail.com/ Red Pill Junkie

    How is this different from Paperman?

    • jonhanson

      Seems more sophisticated. If I remember correctly the Paperman system was more about individual lines whereas this pushes the technology to include full image painting.

      So not totally unprecedented but an interesting evolution.

  • Mister Twister

    In the not-so-distant future, the computers will do all the animating for us. Shortly after that, the writing will be automated as well. In this perfect future, we will have no more art. Or jobs. And then we will all die. I AM HYPED!

    • Gage

      You sound like an extremely out-of-touch old geezer. Just so you know.

  • PETITE pedro

    why they trying to make it something its not?

    • Mister Twister

      Because they want to kill art and replace it with automation?

      Just a wild guess.

      • Jonah Sidhom

        Computer animation is not automated. Good animation, at least.

        And they’re not trying to “make it something it’s not.” It is the combination of two techniques to make something new.

    • Gage

      Why do people try to make their drawings three-dimensional? They’re flat. Why try to make it something it’s not?

  • Dirty Laundry Day

    Can’t help but thinking, Tofu Bacon. am I alone on that one?

  • Mike Scott

    Hmm… I sometimes wonder if these supposed automated approaches end up costing more work and looking not as good as hand-drawn 2D animation. Anyways, worth exploring.

  • jonhanson

    Definitely interesting, but I’d have to see it put into full effect to really judge it. The demo with the skater looks pretty great but the image from Up looks a little too much like those shots where people run footage through Photoshop’s “cartoon” filter and it just sort of comes out a mess.

    But I understand that’s the nature of test footage, here’s hoping we see a short using that technology soon! I’m still waiting for a feature that incorporates some aspects of Paperman’s style.

  • William Bradford

    I think what’s off is that the animation isn’t very good: Now I know that’s not the main point of these demos, but I would think they could put just a bit more sophistication into the animation of the guy skating. I think this technique could open up AMAZING possibilities is used tastefully though, so good on them.

  • Kamikaze

    The beautiful textures Tim Burton and his team accomplished with stop-motion (for example, in The Nightamre Before Christmas) were incredible and I felt they were much more tangible than CG. To me, they’re pretty much like living illustrations.

  • Crispy Walker

    Wooow. “While there is a certain empty attractiveness to rendering in an illustrative style, the surface seductiveness has no influence on either the film or the animation besides reducing the emotional authencity of the artwork by distancing it further from its digital roots. Perhaps that’s the goal, but it seems to me an aesthetic dead-end.”

    Empty attractiveness? Reducing the emotional authenticity? Aesthetic dead-end? That’s one huge hyperbolic statement if I’ve ever read one. If anything, being able to render a computer animated feature in a more illustrative style might bring back some of the emotional authenticity that we lost in the giant shift from 2d to 3d. Animation was easily empathized with because of its illustrative stylization — the films always looked like moving picture-books or comics, and the touch of the hand upon the screen grounds animation in humanity, whether or not the world we’re being shown is our own reality. Emotional authenticity should have nothing to do with the media with which the film was created, unless the media stands in stark contrast to the tone and mood of the story. Good animation is good animation, regardless of the means.

    I don’t see this as an aesthetic dead end either. This is basically what Disney was trying to do with Paper Man, but to an even greater extreme. It’s not a new idea — there’s that french movie The Painting which was done in 3d… but given this kind of technological push behind Pixar’s developments, I think it could have been even better visually. Putting artists to use behind this kind of effect could drastically improve upon the well-worn hyperrealistic 3d environments that we’ve been subjected to in CG animation over the past decade. I think part of the problem in these examples is that many of them don’t look as if they put the workflow model outlined with the final skating elf to use. The Up clip is terrible and looks like a photoshop filter, as does the hatched clip of the dude with the shotgun. However, the little skating elf shows promise.

    My Neighbors the Yamadas is a beautiful film partially due to its washy watercolor rendering — Takahata’s new film looks even more beautiful with those effects pushed even further. I think this might be a great step forward in developing a wider visual vocabulary for CG animation than we have at the moment, and the fact that it’s digital work mimicking the aesthetic of traditional media should in no way diminish from the final film’s “emotional authenticity”. Hell, most illustrators and comic book artists today work digitally with the goal of emulating traditional media in mind to great effect. Anything that could bring about a change from the super-smooth “everything looks the same” look that we have in CG animation right now should be a lauded achievement. We’ve had 15 years of films that all have the same plastic sheen to them — this has the potential to be a huge boon to the viewer’s aesthetic experience of the film, which will ultimately affect their emotional reaction. Seeing a whole film done in a painterly style could be SO beautiful and nostalgic and create such a sense of wonder in a viewer. I don’t think we should be picking this idea apart and be proclaiming it as merely decorative schlock until we’re able to see how it truly can be put to use. I don’t think these examples were great, but it certainly has the potential to bring something new to the stagnant waters of CG animation.

    • paul

      agreed, this is the opening up of animation options to a whole new world, including traditionally-drawn techniques and styles as essentially a faster, cheaper, and more accurate in-betweening method. in that way and otherwise it simply expands the creative flexibility, and hopefully encourages artistic innovation beyond primarily using the computer to generate photo/hyper-realistic renderings.

  • Axolotl

    Yeah! And it will help private individuals with limited funds to produce crappy soulless counterfeit 2-d! Heck, they could make a feature length painterly sterile hack job all by themselves. I don’t foresee anything good coming out of this. No sirree. No.

  • Scott W

    I think the way to consider this isn’t about CG trying to fake painting, but painting being helped by CG. The whole point of the keyframes being painted by hand is to let a real artist have real control. The CG is there to finish the job without the painter’s hands falling off.

    I see this as a way of letting new styles into the realm of animation. Styles that have, until now, been elusive or too complex to render individually on 130,000 moving frames.

  • Taco

    The fundamental argument of 2D (Representational) graphics Vs 3D (Dimensional) graphics is that…
    This is NOT a Pipe (it’s a drawing or painting etc…)
    http://sandiego-smokeshop.com/pix/This-Is-Not-A-Pipe.jpg
    …while…
    this one IS …(for all intent & purpose)… a Pipe
    http://www.exchange3d.com/images/uploads/aff108/Smoking-pipe/001.jpg

    One approach “physically” builds something, while the other simply represents it!!!

    See “Ceci n’est pas une pipe” quote from René Magritte at http://en.wikipedia.org/wiki/Ren%C3%A9_Magritte

    • jonhanson

      If you can’t smoke from it then how is a 3D model for all intents and purposes a pipe?

    • Chris Sobieniak

      I understand you clearly here.

    • archJones

      I’m sorry, but your argument makes no sense. I understand what you’re “trying” to say, but your logic is completely flawed.

      For “all intents and purposes” the 2D drawing IS a pipe too. Because the intent was to make something that looks like a pipe, so when people look at the drawing, they say “hey that’s a pipe”. No different than the CG version. They are BOTH representational, because neither IS a pipe.

      Your argument also loses validity when we start talking about sculptors. So if a sculptor makes a pipe out of clay does that make it a pipe and any less of a form of art because, well, it IS TRUE 3D and you can actually “physically” touch and feel it? No it does not.

      Also, in 3D (CG) you don’t “physically” build the pipe anymore than you push a pencil and brush to render the pipe. You are moving pixels with a mouse or stylus and not touching the form itself, but even if you were refer to the above paragraph about sculptors.

      All kinds of flaws.

      • Taco

        The problem with your argument KevinGrinder is that you’re not even trying to argue a meaning or convey a point here. You’re just arguing over the semantics that I used in trying to put my point or idea across.

        All kinds of flaws in your deductive reasoning right there, buddy… (did you even read my later comments, like the ones in reply to Jonhanson?)

        But maybe I can better express my point more succinctly, so you don’t get stuck at “It’s still not a pipe, even though it’s a fully dimensional 3D object in a dimensional space (virtual or otherwise)” and miss my point. At any rate, I will “try”.

        You’re right in saying that both sculptures & drawings are representations or “proxys” for something, and that they can’t really be the actual thing they are a visual proxy of, now can they? I mean, that’s obvious. We both agree on this, and maybe that’s why I didn’t ad this disclaimer into my initial semantics in my first comment on this subject, because I assumed it was sort of obvious. But the point I was CLEARLY trying to get at, and which you seem to gloss over entirely in your reply, Is the following…

        There is a big big difference between the aspects & creation of a Drawing… VS… the aspects & creation of a Sculpture, (CGI or Clay) Are ya with me here? Do we agree on that sentence so far? Good, I hope so.

        So… What is that difference? THIS is the key question that I was trying to express in my comments & thoughts above. I’m sorry that you found my initial semantic reasoning & crude attempt to express that in words not up to your high standards. Allow me to re-state: A 2D image is only ever a drawn representation of a form from a certain perspective of view within an implied & perceived (ie:drawn) visual space. So… with a drawing, you can only ever see that representation in one particular way, because that is how it was drawn. it DOES NOT exist as anything else beyond that drawing.

        This aspect is inherently DIFFERENT from a 3D Model + environment, looked at via a maya-view-port in cgi-grid-space or your human eye looking at a real word Sculpture sitting on your coffee table. Those things are far more rigid, tangible & rotatable than a mere drawing. There is far less implied & perceived to a sculpture model or digital-rig of a car, than a drawing of said car. And because those dimensional rules are governed by the (for lack of a better word) tangibleness of their dimensional form… they will (in certain ways) lack the versatility & wider range a drawing has, as a representation of something.

        In sort… as a type of representation, Sculptures & Rigs (of any kind), share the same dimensional qualities & are more akin & comparable with that which they are made to represent, than a drawing is & what it is a representation of.

        Thanks for allowing me to try and re-address everything I’ve tried to here Kevin. I hope all of these words & thoughts don’t confuse you further. Cheers.

  • Jack Daniels Old Number 7

    I would rather just have them make traditional animation instead of CGI all together. I am getting so sick of CG to the point where I just don’t enjoy them all together. And plus these CGI movies really advocate casting only stars for almost the entire cast which brings BORING voice acting to the characters instead of actual voice actors who actually know what they are doing.

  • Bob Harper

    I think this test is an arbitrary aesthetic embellishment, but I think the idea can open the possibilities of styles for those who purposely choose that aesthetic to tell a certain kind of story.

    I don’t know who would spend the money to have artists finger paint in oil on glass for a feature or wait years for that kind of film to be finished.

    After all of the complaints of how Pixar and others are so obsessed in trying to create hyper realistic images in lieu of an artistic style, I find it hopeful that they are exploring something different.

  • pooki3

    This will probably be used for Pete Docter’s new pixar movie about the inside of the brain which i hear will have a 2D style

  • Rubber Hose Arms

    Wow, seeing a lot of Luddite hyperbole here. Good grief, people, the simple fact is: they’re experimenting with a different look. That’s it. Remember doing this sort of thing in art school? Yeah, it’s a GOOD thing! Give it a chance. Could lead to some interesting techniques and inspire other developers to try something different.

  • Makar Yeliseykin

    I miss watching cartoons that looked like cartoons, I dont want to have a realistic looking characters and their flowing movements and the details on their table and skin and clothing that will make me go wow.

    I think Fantastic Fox was an example of a “cartoon” that has its own style and when computers will be able to create and replicate that kind of style it will be a great achievement in the world of technology and how it gives us something new. All pixar movies look the same and now I think they are looking into a new direction to make each movie different an unique from one another.

  • CG Animator

    So… traditionalists complain about CG because it’s “too realistic” in the rendering style… and then Pixar comes along and does a test with a non-realistic painterly rendering style, and they still complain because… well… it’s CG and CG is eeeevvvviill.

    Well guys, looks like we can’t win either way. Oh well, if you can’t beat ‘em, join em:

    Blah blah blah “CG animators aren’t artists” blah blah blah “Outdated CG animator tech-nerd stereotype from 1987″ blah blah blah “CG animators can’t/don’t/are not allowed to draw” blah blah blah “it’s all automated” blah blah blah rant rant grumble grumble.

    • UsaMiKo

      I’ve got an admitted bias for 2d animation, but I don’t think that 3d animation is evil…

      … I think that Hollywood is.

  • Justin Miller

    There seems to be no loss in emotional authenticity in digital art that mimics traditional media. What informs your conclusion that this is so certainly the case when applied to cgi rendered media that mimics traditional media? This smacks of a conclusion in search of an imagined problem.

  • http://otherthings.com cassidycurtis

    It’s nice to see non-photorealistic rendering get some exposure here. But, Amid, I think you’re making the mistake of critiquing a research test example video as if it were a polished film. The two serve very different purposes. The video says “here is what CAN be done”. It’s up to the filmmakers to decide what SHOULD be done.

    NPR is not, in general, an attempt to “change the style of CGI to mimic traditional media”. It is simply a set of tools that enable filmmakers to control the look of their film down to an unprecedented level of detail. For example: when you paint on glass, you don’t get to choose whether your brushwork should move smoothly or erratically. It always moves erratically. The option of smoothness is unavailable to you. With NPR techniques like this one, you can now make that aesthetic choice. Does this mean every animator will always choose smooth movement for every moment? Absolutely not. Only a very naive art director would simply slap a painterly look on top of an already-finished piece of CG animation as you suggest.

    The tone of your post reveals a kind of essentialism that I find objectionable on both artistic and technical grounds. You make false, ignorant assumptions about the limitations of CGI, and then when the medium transcends your imagined limitations, you cry “aesthetic discord!” Well, you can’t have it both ways. NPR is part of CGI, and has been for over 25 years. Deal with it.

    • AmidAmidi

      Cassidy, You wrote, “When you paint on glass, you don’t get to choose whether your brushwork should move smoothly or erratically. It always moves erratically. The option of smoothness is unavailable to you.” To that, I say absolutely. Of course. We are in total agreement that the unique properties of a chosen material impact the final image. The issue is that NPR (even more than photorealistic rendering) separates the material from the image.

      In most art, you have an interplay between structure and texture. Take the drawing here by John Constable, there is no way that you can take the two things apart because structure and texture work in harmony as a whole.

      In CG, it is entirely possible to create a synthetic representation of this image by constructing the structure first and adding the texture atop. It’s a two-step process that will inevitably be poorer than the original, because unlike pen-and-ink, the CG texture does not inform the structure.

      It creates, as you say, smoothness—smoothness that recreates the effect of a material while eliminating the imperfections that result from the authentic material’s behavior on the image. The problem with layering, say for example, a pen-and-ink style atop CG animation is that the animation would not actually look or move as such if it had been created with pen-and-ink. The unique properties of ink dictates where you place your strokes, how you construct your forms, and why you time movements a certain way. All of that is lost if something is post-processed with ‘pen-and-ink’ over movement which has already been generated. “I paint forms as I think them, not as I see them,” Picasso said. With the current process Pixar has devised, the artist cannot think. He can only paint over what is directly in front of him.

      When the material is separated from structure and movement, it creates a level of artifice that’s difficult for me to penetrate as a viewer. As I watch “Paperman” or “Fishing,” I see the attempt to replicate an aesthetic, but absent the properties and behaviors that are germane to the graphic material. The film, thus, becomes aesthetically discordant, a graphic cipher in which the movement doesn’t match the material that the filmmaker wants us to believe was used in its creation.

      You’re entitled to tell me I’m wrong for feeling this way, and I’m entitled to feel this way until I see something that changes my opinion. We can all agree that CG is still in an extremely primitive state. Perhaps at some point in the future, artists will be able to model and animate with the materials they’ve chosen, and their modeling and animation will be dynamically influenced by the choice of those materials. Better yet, perhaps CG will pursue the discovery of its own erratic aesthetic instead of smoothing out other forms.

    • AmidAmidi

      Cassidy – You write, “When you paint on glass, you don’t get to choose whether your brushwork should move smoothly or erratically. It always moves erratically. The option of smoothness is unavailable to you.” To that, I say absolutely, of course. We are in total agreement that the unique properties of a chosen material impact the final image. The issue is that NPR (even more than photorealistic rendering) separates the material from the image.

      In most art, you have an interplay between structure and texture. Take the drawing here by John Constable, there is no way that you can take the two things apart because structure and texture work in harmony as a whole.

      In CG, it is entirely possible to create a synthetic representation of this image by constructing the structure first and adding the texture atop. It’s a two-step process that will inevitably be poorer than the original, because unlike pen-and-ink, the CG texture does not inform the structure.

      It creates, as you say, smoothness—smoothness that recreates the effect of a material while eliminating the imperfections that result from the authentic material’s behavior on the image. The problem with layering, say for example, a pen-and-ink style atop CG animation is that the animation would not actually look or move as such if it had been created with pen-and-ink. The unique properties of ink dictates where you place your strokes, how you construct your forms, and why you time movements a certain way. All of that is lost if something is post-processed with ‘pen-and-ink’ over movement which has already been generated. “I paint forms as I think them, not as I see them,” Picasso said. With the current process Pixar has devised, the artist cannot think. He can only paint over what is directly in front of him.

      When the material is separated from structure and movement, it creates a level of artifice that’s difficult for me to penetrate as a viewer. As I watch “Paperman” or “Fishing,” I see the attempt to replicate an aesthetic, but absent the properties and behaviors that are germane to the graphic material. The film, thus, becomes aesthetically discordant, a graphic cipher in which the movement doesn’t match the material that the filmmaker wants us to believe was used in its creation.

      You’re entitled to tell me I’m wrong for feeling this way, and I’m entitled to feel this way until I see something that changes my opinion. We can all agree that CG is still in an extremely primitive state. Perhaps at some point in the future, artists will be able to model and animate with the materials they’ve chosen, and their modeling and animation will be dynamically influenced by the choice of those materials. Better yet, perhaps CG will pursue the discovery of its own erratic aesthetic instead of smoothing out other forms.

      • Draško Ivezić

        Amid, you are absolutely right when you are talking of creating the original images as you showed in your example, but we are talking about animation industry here.

        So I would like to know – how is by your opinion this method any different from traditional cel animation when you have the army of cleaning artists trying to accomplish desired look from the production bible on the top of the hand drawn rough animation? When you look through the pipeline of any Ghibli movie there are tones of wonderful sketches by maestro Miyazaki, but at the end there is labor put into everything look similar and clean, and there is a reason for that, because Miyazaki original style is impossible to mimic on such a big scale, so they have to compromise and design the final look to fit to the production pipeline.

        One thing is stylistic choice which this method is offering, but another question is the question of the labor and executing the goal for given cost and time to make your product survive in this highly competitive market. You can discuss production of Alexandr Petrov (which is very often financed by subsidies) where is lot of labor put into look and feel of painted surface, but it is also known that he is heavily into using rotoscoping for producing such images, he said himself many times. So, the only issue I see here is the labor which by this technique will be replaced by computers, but creative decisions are still into human hands, such as story and stylistic ones.

        I would not be surprised, after watching Her (2013, Spike Jonze) that in the near future directors will sit with Oculus Rift, let’s say, and as theatre director will give some commands to their AI cast, after that it will be just the execution of the final image, which is then replacing more human labor with computer. This is realistic scenario, and I can cry in advance, but what makes us human storytellers my guess no computer can learn to mimic, at least I hope so. Until we end up on some needle like Keanu in Matrix, let’s enjoy while we still have the saying in the creative work.

        • peter wassink

          Funny you mention “Her”. The work of David OReilly, who animated the game sequence in that film, is the pinnacle of CGI avantgarde.
          He is using the CGI technique not for hyperrealism nor to mimic traditional techniques. Quote by OReilly:” I … really love 3D. I still feel it’s at its earliest stage and I get excited about doing ideas that only work in that medium.” The point i think Amid is making is that this painterly dressing up of CGI animation doesn’t seem to be doing that at all.

      • http://otherthings.com cassidycurtis

        Amid, with respect, your understanding of NPR techniques is as shallow as your knowledge of art history is deep. A technique like the one in the video above would indeed be woefully inappropriate for reproducing a John Constable drawing, and only an amateur would try to use it for that purpose. But NPR is not limited to just that one technique (advected textures that track surface movement) or even that family of techniques (things that track surface movements) or input models (surfaces).

        If you’d like to educate yourself about NPR before talking about it again, you could start by reading my 1999 SIGGRAPH course notes on the subject (http://otherthings.com/uw/npr/early_version.html). Then, go back and survey the last fourteen years’ worth of proceedings of the NPAR symposium (http://expressive2014.richardt.name/NPAR/Home). There’s a wide range of research topics in there that should give you a sense of the breadth of what is actually possible.

  • jonhanson

    You could do that now, you just have to be willing to draw a hell of a lot.

    If you’re asking whether it will ever be relatively affordable, maybe but I doubt it, especially the way you describe it.

  • eztempo

    So this is the Instagram of animation, huh.

  • zoe

    Yet another innovation for keeping labor costs down via automation of things which could previously only be done 100% by hand. Personnel is the biggest expense in any studio’s budget. Anything that makes it possible to reduce skilled labor will make the executives lick their chops.

  • llt

    putting everything though a default photoshop filter looks great,doesn’t it.

  • fbt

    I think animation studios will start making 2D animation when every one in the CGI animation industry stop under-appreciating paper and start appreciating the fact animation started by flipping a bunch of papers and not by clicking a few buttons.

    So it’s never going to be you who decides.

    • dave

      i’m sorry, flipping a few buttons?? it seems to me you have little to no understanding of the time, skill, and patience it takes to do CGI animation

  • Fingleberry

    This is really cool. Reminds me of when Square Enix’s Visual Works was
    experimenting with a painterly look back in the early 2000s.

    http://www.youtube.com/watch?v=lLnPULAf4MY&feature=player_detailpage#t=583

  • vincenzosz

    No, but the animation industry is one of making money. So this is a technology that can save time, maintain the CG element that seems to be needed to attract an audience, while still bringing back that appeal and aesthetic of hand drawn (or painted, etc) animation.

  • CG Animator

    I do a lot of Flash “puppet” work as well as 3D so I know exactly how you feel Bob lol. Love your work btw.

  • CG Animator

    I have nothing agaist traditional animators, believe me. I just hate the additude a lot of them have against CG. Like CG is somehow not “real” animation because we animate with a mouse or a tablet instead of a pencil.

  • Lucky Jim

    Wow. Imagine what this thing will look after 20 years of continual use and refinement.

  • Googamp32

    If it makes you feel any better, I love Flash animation! It’s CGI I hate!

    • Mike L

      Flash is also CGI but whatever rows your boat.

  • Steve

    How are we not talking about Batz in tandem with this demo? https://vimeo.com/67726825

    http://www.kickstarter.com/projects/687407823/batz-0

    I think the style of this short is pretty 2D. It looks fucking awesome!

  • George Comerci

    Just watching this video got me excited! I like the look, and the fact that it combines the smooth, detailed movement of CG and combines it with something that looks like it’s straight out of a painting. Like what Paperman did. As for animation studios using this technique in the future…? I wouldn’t want to see it for every movie, but if a film is going for a certain style then I would love to see it.

  • Taco

    I know & understand Magritte’s point & where he was coming from. I have read plenty about the man & his art. I am CLEARLY not using his quote in the direct context that he said it in. I am using it to help illustrate the differences between a visual representation of something (like a drawing), Vs a fully dimensional re-creation of something (like a 3D-GCI Model). Something that Magritte was not trying to do. Honestly Leosch, If you can’t take information or a statement from somewhere and use it as a tool to help illustrate what YOU are trying to convey, you’ll never say anything of interest.

  • CaJd554

    This has been done far better by others. YEARS ago. But then, technologically, Pixar was only once ahead of the curve. I hope they drop their jones for technology and focus on better storytelling. Maybe then their technology will catch up. It’s what’s only worked before.

  • jonhanson

    Just having some fun. I know my comment may seem pedantic, but Magritte’s painting probably seemed similarly pedantic to people at the time.

    You really get to the heart of the power of CG, and it’s restrictions.

    • Taco

      Thanks jonhanson, I appreciate you or anyone taking the time to read my attempts at reasoning & help evaluate my logic. If what I’ve said makes sense, please feel free to use it yourself. Personally I think this whole 3D Vs 2D “Civil War” is just a bunch of baloney! Some people just don’t like using 3D programs like Maya & the spline-spaghetti-box to animate their characters and much prefer the process of Drawing & designing. But most just wanna animate and… lets face it, there are currently far more job opportunities to be an “Animator” in Films, TV or Games if you’re a wiz at Maya & or Flash. and @CG Animator: “I just hate the attitude a lot of them have against CG. Like CG is somehow not “real” animation because we animate with a mouse or a tablet instead of a pencil.” If any “traditional” artist ever says anything like that, then you have my full support to go break their nose & all their pencils. Likewise, if any 3D animator tells me that “2D is Dead & I should just crawl into a hole & cry!”, I hope that I have YOUR full support to break that person’s nose & their keyboard! Apparently 85% of all arguments are over a marginal differing in Semantics and the fact that we can’t entertain every persons perspective all at once. (i’m not sure how true 85% is, but I’m sure it’s atleast 50% of all arguments, especially on the internet.)

  • IJK

    How exactly does a 3D model become less clean? There’s not exactly sketch marks you can preserve over it.

  • Samuel

    Ok, ok, ok…
    Frankly Pixar go on ! Continue like this !
    Harmonizer on vox to sing right, 3D to draw right, CNC to sculpt right…our world will sure look like paradise, paradise for the nazis.
    I don’t care if it’s made with a computer or not but show me just one 3D shit that has a soul. I mean a real one, no something that has a soul effect on it.
    Amount of work, benefit, war to prove if it’s artistic or not, who works the most : bullshit !

    I don’t want to wake up and realize that the woman I slept with is a robot.
    “Wow !! incredible that was so realistic!”

    People all over the planet are sick and so stupid that they don’t realize what they are doing : they are creating a huge hologram, they are travestying everything and want me to tell there is no problem with it….
    even get angry if I don’t want to be agree.
    There is a huge PROBLEM !!! wake up !!!

    Put your ego at side and ask yourself what you are doing and what is your responsibility is this shiny world.
    You really want everything to be a lie ?

    Go on ! but don’t ask me to be part of it and to be gentle to not punch you in the face every time I will have the occasion to.

    • Gage

      ….

      I am at a loss for words. I’m going to continue on my day believing this comment was satire, because I don’t know how to process it otherwise.

  • Steven Bowser

    I think the intent is to take the best from hand-drawn and the best from CG and put them together. With CG animation you can have dynimic camera movements and complex simulations and animation, but with hand-drawn you can have a more intimate and personal aesthetic and a more flexible style. If you can improve the rigs enough and infuse the animation with hand-drawn/painted elements then CG can have more of that personal touch that makes it so attractive.
    I think it worked wonderfully with Paperman, and I’d love to see it taken even farther as long as the focus is left on the storytelling. If the visuals are too flashy then it could be distracting and remove the suspension of disbelief.

  • Animator606432

    That’s what I don’t get. Making 3D animation looks less “plastic like” and bringing textures in is cool. But if it’s just to make it look like 2-D animation…why bother? That’s sort of a waste of technology to be honest. We already have a process to achieve that look,

  • Animator606432

    I disagree. I can see hand drawn making a comeback, it’s just that nobody wants to put the effort in to make one that could be successful. (Even thought PaTF was successful). If stop motion animation can become popular again, what can’t traditional animation do the same? As I said before, audiences like variety. There shouldn’t be one dominant form of animation, there should be several different ones.
    On top of all that, the Europeans and the Japanese, love traditional animation. They have CG but there is PLENTY of traditional animation that goes in their theaters.

  • IamSam

    That is nothing special. Somebody already figured it out without creating anything special or reprogramming and quick rendering.

    http://vimeo.com/5660045

    • schwarzgrau

      This one fakes brush-strokes through motion blur. In Pixars way you could actually use real paint, since it kinda projects an painted image on the CG-model.

    • George_Cliff

      While the look bears some similarity and seems to be attempting to achieve the same end, there is simply no comparing the quality of this effect to that of Pixar’s. This effect is vague and muddied like everything is covered in gauze. The effect is far too conspicous and my eyes tired watching the 3 second sample clips they showed. This would not hold up for the duration of an animated short much less a feature film.

  • schwarzgrau

    No TeamCerf uses a different approach. They paint their textures by hand in Photoshop.
    Pixar rotoscopes every tenth frame and use the CG-animation to interpolate the missing frames.

  • archJones

    Paper and pencil, made “Home on the Range”, “The King and I” and countless other horrible 2D animations. Googamp32, that means you should hate 2D too.

    Fact is there are good and bad in both, so saying you hate one because of a bad film has nothing to do with whether or not it was 2D or 3D. Those films were not bad because of the technique used, but many factors, starting with story.

    There are some ridiculous points on here. I mean I’m sure old masters only ever practiced or “liked” only one form of art. A sculptor never draws or paints, right? You’ll never catch a painter touching clay- what a sin. Just ridiculous.

    I love all animation. I don’t care what form it is in. I appreciate the artform and love to see what limits it can be pushed, from pixelation, stop-motion, sand to CG. The artform is great, but that does not mean all works are great.

  • Stubbs

    This style has been around for some time. Funny that the big guys now start to adopt it. in a very technical way…

    Example:
    https://vimeo.com/41656414